Time series analysis of macroeconomic and financial variables requires a deep understanding of many econometric pitfalls if an empirical researcher hopes to avoid making spurious inferences. This understanding is the hallmark of Charles Nelson's research over four decades and it develops out of a healthy skepticism about “conventional wisdom,” yet a pragmatic belief that, despite the econometric hurdles, it is possible to learn from data. The papers in this special issue build on Charles Nelson's research legacy to address many important empirical questions related to business cycles, financial markets, and inflation, always with respect for the data, but wary of spurious inferences.

This paper employs the state-space model to reexamine the fundamental issue in finance of whether it is the expected returns or the expected dividends growth that is primarily responsible for stock price variations. We use Bayesian methods to show that there is a substantial uncertainty about the contributions of expected returns and expected dividends to fluctuations in the price–dividend ratio when the aggregate returns and dividends data are used. The substantial uncertainty of the contributions results from the model being weakly identified. Our finding challenges the notion long held in the existing literature that it is the expected returns that contribute most to price–dividend variations.

We analyze the forecasting performance of small mixed-frequency factor models when the observed variables share stochastic trends. The indicators are observed at various frequencies and are tied together by cointegration so that valuable high-frequency information is passed to low-frequency series through the common factors. Differencing the data breaks the cointegrating link among the series and some of the signal leaks out to the idiosyncratic components, which do not contribute to the transfer of information among indicators. We find that allowing for common trends improves forecasting performance over a stationary factor model based on differenced data. The “common-trends factor model” outperforms the stationary factor model at all analyzed forecast horizons. Our results demonstrate that when mixed-frequency variables are cointegrated, modeling common stochastic trends improves forecasts.

A well-documented property of the Beveridge–Nelson trend–cycle decomposition is the perfect negative correlation between trend and cycle innovations. We show how this may be consistent with a structural model where permanent innovations enter the cycle or transitory innovations enter the trend, and that identification restrictions are necessary to make this structural distinction. A reduced-form unrestricted version is compatible with either option, but cannot distinguish which is relevant. We discuss economic interpretations and implications using U.S. real GDP data.

Does excluding food and energy prices from the Consumer Price Index (CPI) produce a measure that better captures permanent price changes? To examine this question, we decompose CPI inflation and “core” inflation into their permanent and transitory components, using a correlated unobserved-components model. The stationarity of inflation may be time-varying, so we examine the performance of the core measure of inflation separately for periods in which inflation is I(1) and I(0). For a period in which inflation appears to be I(1), we find that core inflation and the permanent component of overall inflation are closely related, but there are some caveats. For a period in which inflation appears to be I(0), we decompose the core and overall price levels and find that the permanent component of the core CPI is much more volatile than the actual core series and that the core excludes volatile permanent shocks to the overall price level.

This paper investigates the impact of institutions on the dollarization of the domestic banking system using a unique policy experiment: the process of accession of countries to the European Union (EU). Using a dynamic factor model, we decompose fluctuations in financial dollarization for 24 transition economies into a common factor, an EU factor, a non-EU factor, and country-specific factors. The EU factor, which proxies for improvements in institutions under the set criteria for eventual membership, reveals the importance of institutions for the extent of financial dollarization over time. The results also indicate the asymmetric impact of improved institutions on the domestic bank's balance sheets by inducing higher loan dollarization and lower deposit dollarization. The relative importance of the EU factor to the financial dollarization of a country is associated to the degree of comovement of its business cycle with that of the EU.

This paper explores the hypothesis that the sources of economic and financial crises differ from those of noncrisis business cycle fluctuations. We employ Markov-switching Bayesian vector autoregressions (MS-BVARs) to gather evidence about the hypothesis on a long annual U.S. sample running from 1890 to 2010. The sample covers several episodes useful for understanding U.S. economic and financial history, which generate variation in the data that aids in identifying credit supply and demand shocks. We identify these shocks within MS-BVARs by tying credit supply and demand movements to inside money and its intertemporal price. The model space is limited to stochastic volatility (SV) in the errors of the MS-BVARs. Of the 15 MS-BVARs estimated, the data favor a MS-BVAR in which economic and financial crises and noncrisis business cycle regimes recur throughout the long annual sample. The best-fitting MS-BVAR also isolates SV regimes in which shocks to inside money dominate aggregate fluctuations.

We investigate the importance of trend inflation and the real-activity gap in explaining inflation in G7 countries since 1960. Our analysis is based on a bivariate unobserved components model of inflation and unemployment in which inflation is decomposed into a stochastic trend and a transitory component. As in recent implementations of the New Keynesian Phillips Curve, it is the transitory component of inflation, or “inflation gap,” that is driven by the real-activity gap, which we measure as the deviation of unemployment from its natural rate. We find that both trend inflation and the inflation gap have been consistent and substantial determinants of inflation at business cycle horizons for all G7 countries since 1960. Also, the real-activity gap explains a large fraction of the variation in the inflation gap for each country. These results provide empirical support for the New Keynesian Phillips Curve augmented with trend inflation.

Early research on the Taylor rule typically divided the data exogenously into pre-Volcker and Volcker–Greenspan subsamples. We contribute to the recent trend of endogenizing changes in monetary policy by estimating a real-time forward-looking Taylor rule with endogenous Markov switching coefficients and variance. The response of the interest rate to inflation is regime-dependent, with the pre- and post-Volcker samples containing monetary regimes where the Fed did and did not follow the Taylor principle. Although the Fed consistently adhered to the Taylor principle before 1973 and after 1984, it followed the Taylor principle from 1975 to 1979 and did not follow the Taylor principle from 1980 to 1984. We also find that the Fed only responded to real economic activity during the states in which the Taylor principle held. Our results are consistent with the idea that exogenously dividing postwar monetary policy into pre- and post-Volcker samples is misleading. The greatest qualitative difference between our results and recent research employing time-varying parameters is that we find that the Fed did not adhere to the Taylor principle during most of Paul Volcker's tenure, a finding that accords with the historical record of monetary policy.